Embodiments relate to an optically controlled electrical-switch device based upon carbon nanotubes and to an electrical-switch system using the switch device.
In the last few years, the considerable success of CMOS technology has been determined fundamentally by the possibility of constantly reducing the dimensions of electronic devices. In fact, this technology follows the so-called Moore's law, according to which the number of transistors that can be obtained on an integrated circuit and, consequently, the speed of calculation should double in a time range of between 18 and 24 months.
However, it is a common conviction that conventional silicon micro-electronics cannot continue indefinitely to follow this law in so far as sooner or later physical limits that prevent current circuits from functioning in a reliable way at nanometric dimensions will certainly be reached, while at the same time an exponential increase in production costs will render any further increase in the levels of integration prohibitive. By increasing the density of the electronic devices on a chip, in fact, phenomena such as the need to dissipate the heat generated by such dense circuits and the transition from the classic behavior to the quantum behavior of charge carriers will considerably slow down progress.
In particular, thanks to the use of lithographic techniques, there have currently been reached dimensions of the order of 100 nm. Notwithstanding the rapid progress achieved in the current process of scale integration, current technology is difficult to scale further below these critical dimensions. In fact, once the critical dimensions have been reached, the small electrical currents that carry the information are transferred uncontrollably from one device to the other. In particular, when quantum effects start to become important, the transistors tend to lose the electrons that represent the information, so that it becomes difficult to maintain them in the original state. It is envisaged that, below the dimensions indicated of 100 nm, these difficulties are likely to become important.
The need to solve the above problems has forced research in the direction of the study of new technologies based upon the use of organic materials that can replace, altogether or in part, silicon in the construction of electronic devices.
Molecular electronics has the potentiality for overcoming the limits of silicon technology in so far as it is possible to fabricate single molecule devices that organize themselves in parallel by means of self-assembly techniques, which are also economically advantageous.
The need has thus arisen to explore the possibility of passing from current assembly technologies of the top-down type, whereby it is possible to reach the desired dimensions with successive removals of a macroscopic amount of a material, to technologies of the bottom-up type, whereby it is possible to make, and subsequently assemble, nanometric components starting from individual atoms or molecules, i.e., ones in which the devices involved in handling and retaining the data are molecules arranged and interconnected so as to form a circuit.
Amongst the different molecular structures studied, carbon nanotubes (CNTs) have aroused an enormous interest owing to their extraordinary physical properties. For a detailed treatment, see for example the article “Carbon Nanotube-Based Nonvolatile Random Access Memory for Molecular Computing”, Science, Volume 289 (5476), Jul. 7, 2000, 94-97.
It is known the property of carbon atoms to organize themselves into different structures, giving rise to materials of different forms. In fact, diamond is made up of carbon atoms organized in tetrahedrons, while graphite is made up of carbon atoms organized in planar structures. These two allotropic forms, albeit arising from the same type of atoms, exhibit structural properties (hardness, elasticity, friction) and functional properties (electrical conductivity, color, etc.) that are very different and frequently opposite.
The structural characteristics, such as hardness and refractoriness, of graphite and diamond render it difficult to implement, at nanometric scales, an on-device approach of the top-down type. Instead, an approach of the bottom-up type is made possible by the use of another allotropic form of carbon, namely fullerene.
Belonging to the family of the fullerenes is C60, also known as buckyball, which has a molecular structure having the shape of a polyhedral cage, made up of pentagons and hexagons. The structures of fullerenes, which develop in the form of long cylinders, rather than in the form of spheres, are called nanotubes. Their length (several microns) can be thousands of times larger than their diameter (just a few nanometers). Furthermore, using known techniques of molecular synthesis, there have been observed, in the laboratory, single-walled cylindrical structures, or single-walled nanotubes (SWNTs), having a diameter of 1-2 nm, and multiple-walled structures, or multiple-walled nanotubes (MWNTs), i.e., ones formed by coaxial cylinders, with diameters of some tens of nanometers.
Carbon nanotubes are organic molecules made up of a number of carbon atoms interconnected in a cylindrical structure, which are characterized by a low weight and have exceptional elastic properties that render them extremely hard but also capable of undergoing large deformations without breaking. Thanks to their exceptional mechanical properties and to their capacity of conducting electrical charges, carbon nanotubes, in so far as they can be configured both as conductors and as semiconductors, are suited for forming the components of a new class of nanometric electronic devices. In particular, they are believed to play a primary role in the development of molecular electronics on account of the fact that, thanks to their lateral dimensions of the order of the nanometer and their electrical conduction properties, they behave as quantum conductors of nanometric dimensions (“quantum nanowires”).
Carbon nanotubes have different shapes that can be described by a vector, referred to as chiral vector C, as illustrated in FIG. 1.
In particular, in geometrical terms, a carbon nanotube can be obtained from a sheet of graphite, by “cutting” it along lines (dashed lines in FIG. 1) perpendicular to the chiral vector and by “rolling” it in the direction of the chiral vector itself. In this way, a cylinder of diameter d=|C|/π is formed.
The chiral vector consequently defines the type of winding to which the individual sheet of graphite is subjected in order to give rise to a particular carbon nanotube. When the sheet of graphite is wound to form the cylindrical part of the carbon nanotube, the ends of the chiral vector are joined. The chiral vector hence represents a circumference of the circular section of the nanotube.
The chiral vector C can be set in relation to two unit vectors a1 and a2 that define the lattice of the planes in the graphite, by means of two indices n and m, according to the following equation:C=n·a1+m·a2 
Linked to the indices n and m are an angle φ, referred to as chiral angle, and
      ϕ    =          arc      ⁢                          ⁢      cos      ⁢              ⌊                              3                    ⁢                                    (                              n                +                m                            )                        /            2                    ⁢                                    (                                                n                  2                                +                                  m                  2                                +                nm                            )                                      ⌋                  d    =                  a        π            ⁢                                    n            2                    +                      m            2                    +          nm                    the diameter d of the carbon nanotube according to the following equations:
The values of the indices n and m define the chirality of the carbon nanotube, which is the state of the carbon nanotube itself, which differs according to the way in which the hexagons of the graphite arrange themselves in forming the cylindrical structure. The chirality of a carbon nanotube is thus given by the pair of integer indices (n, m) and determines the structural characteristics and, consequently, the electrical conduction properties of a carbon nanotube. In particular, in relation to the structure, nanotubes that have the indices n and m equal, i.e., nanotubes (n, n), are referred to as armchair nanotubes on account of the arrangement of the hexagons of graphite with respect to the axis of the carbon nanotube itself; nanotubes in which one of the two indices is zero (n, 0) are referred to as zigzag nanotubes; nanotubes for which the relation m=0 or else n=m is valid are referred to as achiral nanotubes; while nanotubes with different indices are in general referred to as chiral nanotubes.
The chirality conditions the conductance of the carbon nanotube, its density, its lattice structure, and other properties. The chiral indices can, in principle, be obtained experimentally, by measuring the chiral angle φ and the diameter d of the carbon nanotube with a transmission electron microscope (TEM) or with a scanning tunneling microscope (STM).
Furthermore, according to their chirality the nanotubes can be metallic nanotubes or semiconductor nanotubes. In fact, nanotubes whose chirality indices satisfy the following relation:n−m=3·l l=0,1,2, . . . .
are metallic and, hence, conductors; all the others have a nonzero bandgap and, consequently, behave as semiconductors. Armchair nanotubes are metallic.
The fundamental bandgap of a semiconductor carbon nanotube depends upon the diameter d of the carbon nanotube, on the basis of the following relation:Egap=2y0acc/d 
where y0 is the binding energy of the carbon atoms, and acc is the distance between two neighboring carbon atoms.
Consequently, by appropriately modifying the chirality of the carbon nanotube and, hence, its diameter, it is possible to modulate its bandgap. The two different geometrical structures of the molecule (i.e., the initial one and the modified one) can thus represent two stable states.
Carbon nanotubes can be produced in macroscopic amounts using different techniques: laser ablation, arc discharge, or else chemical-vapor deposition (CVD). For a more detailed treatment as regards the latter technique see for example H. M. Cheng et al., Appl. Phys. Lett. 72, 3282 (1998).
In particular, the latter technique is compatible with the methods used in the micro-electronics industry and enables nanotubes to be grown on substrate. Using the various techniques, it has been found that the carbon nanotube that can be produced in the largest quantity is the carbon nanotube (10, 10).
As has been said, carbon nanotubes constitute a way for responding to the need to reduce the dimensions of devices in integrated circuits. In fact, by means of the versatile molecules, the path has been opened to the construction of molecular transistors.
In the last few years, different configurations of field-effect transistors have been proposed that use carbon nanotubes with semiconductor properties as channels for the transport of electrical charges. Some research groups (R. Martel et al. of the IBM Research Division, Chongwu Zhou et al. of the University of Southern California, etc.) have obtained a so-called “back-gate” configuration, illustrated, in schematic cross section, in FIG. 2, in which the substrate functions as gate of the device. The configuration renders, however, impossible the integration of a high number of devices on one and the same chip. In fact, in this case it would be necessary to apply the same gate voltage to all the transistors on the chip. For a more detailed treatment of the subject, see R. Martel, T. Schmidt, H. R. Shea, T. Hertel, and Ph. Avouris, App. Phys. Lett. 73, 2447 (1998) and C. Zhou, J. Kong, E. Yenilmez, H. Dai, Science, 290, 1552 (2000).
Subsequently, in 1998 researchers of the Dekker group of Delft University of Technology developed carbon-nanotube field-effect transistors (CNT-FETs) using an innovative configuration referred to as “local gate” configuration, illustrated in FIG. 3, which enables integration of a large number of devices on a single chip. For a more detailed treatment of the subject see A. Bachtold, P. Hadley, T. Nakanishi, C. Dekker, Science, 294, 1317 (2001).
Broadly speaking, in the CNT-FETs the channel is constituted by a carbon nanotube functioning at room temperature, and the local gate is insulated from the carbon nanotube by means of an oxide layer of just a few nanometers in thickness. In particular, two gold electrodes were deposited, which function as source and drain, on a silicon oxide substrate grown on silicon, which functions as gate, and these two electrodes were coupled with a single-walled nanotube (SWNT), which functions as channel. The current-voltage characteristic of this three-terminal device was measured and it was verified that it respected the characteristic of a field-effect transistor.
In particular, FIG. 3 illustrates a schematic cross section of the local-gate CNT-FET. The gate in the device is constituted by an aluminium wire with an insulating layer of Al2O3, which separates it from the carbon nanotube. The semiconductor carbon nanotube is set in electrical contact with the two gold electrodes. The thickness of the Al2O3 layer (a few nanometers) is much smaller than the separation between the electrodes (˜100 nm). This determines an excellent capacitive coupling between the gate and the carbon nanotube, the consequences of which result in a gain in voltage greater than 10 and in a wide range of the output signal. It is possible to design various aluminium local gates in such a way that each will address a different CNT-FET.
Formation of circuits comprising CNT-FETs is articulated in three fundamental steps. In the first step, the gate is obtained by delineating the aluminium pattern via electron-beam lithography (e-beam lithography) on an oxidized silicon wafer. The layer of insulating material is represented by an oxide grown by exposing the specimen to air. The thickness of the layer cannot be determined with great precision but is in the order of a few nanometers. The second step consists, instead, in dispersing single-walled nanotubes (SWNTs), previously produced by laser ablation, on the wafer starting from a suspension of dichloroethane. The nanotubes of appropriate diameter (˜1 nm) are selected and positioned on the top part of the aluminium gate. An alternative technique envisages the growth in situ of the nanotubes using the chemical-vapor-deposition (CVD) technique assisted by orienting electrical fields (for a detailed treatment of the subject see Y. Zhang, A. Chang, J. Cao, Q. Wang, W. Kim, Y. Li, N. Morris, E. Yenilmez, J. Kong, H. Dai, App. Phys. Lett. 79, 3155 (2001)). Finally, the last step consists in the formation of electrodes and interconnections via electron-beam lithography, evaporating gold directly on the carbon nanotube without intermediate adhesion layers.
The CNT-FET illustrated in FIG. 3 is a p-type device functioning by enrichment since it is possible to obtain a marked modulation of the current through the FET by applying a small negative voltage to the gate. Furthermore, by acting on the gate voltage it is possible to vary the carrier concentration of the carbon nanotube, up to reversing its polarity from the p-type regime to the n-type regime.
FIG. 4 illustrates the current/source-drain voltage characteristics (I/Vsd) of the transistor measured at room temperature for different values of the gate voltage (Vg). As may be noted, the trend of the curves is typical of traditional FETs with finite values of the current, when the gate voltage is negative and smaller than the threshold voltage Vt (Vt˜−1.0 V).
The same research group mentioned above proposed, moreover, the possibility of making some elementary logic circuits based upon CNT-FETs. The applications (OR gates, AND gates, NOT gates, SRAMs) were obtained using the resistor-transistor logic scheme and forming the CNT-FETs on the same chip.
FIG. 5 gives the input-output transfer characteristics of a logic inverter with carbon-nanotube FETs and a pull-up resistor on the outside the chip with the value of 100 MΩ. The researchers identified the voltage of −1.5 V as an appropriate value for logic applications (logic 0=0 V, logic 1=−1.5 V).
When the input of the logic inverter assumes the logic value “1” Vin=−1.5 V), the negative gate voltage induces a movement of electron holes in the carbon nanotube giving it a resistance that is considerably smaller than the pull-up resistance, and this pushes the output to the logic value “0” (Vout=0 V). When the input assumes instead the logic value “0” (Vin=0 V), the carbon nanotube is not conductive and the output assumes the logic value “1” (Vout=1.5 V).
Recently, some studies have demonstrated the possibility of making molecular devices with non-electrical control using carbon nanotubes. In particular, an example of optically controlled molecular device based upon carbon nanotubes has been proposed by a research group of the University of California. For a detailed treatment of the subject see D. W. Steuerman, A. Star, R. Narizzano, H. Choi, R. S. Ries, C. Nicolini, J. F. Stoddart, J. R. Heath, J. Phys. Chem. B 106, 3124 (2002).
The above work investigated the interactions between single-walled carbon nanotubes (SWNTs) and two types of polymers:    PmPV (poly{(m-phenylene-vinylene)-co-[(2,5-dioctyloxy-p-phenylene)vinylene]}), and    PPyPV (poly{(2,6-pyridinylene-vinylene)-co-[(2,5-dioctyloxy-p-phenylene)vinylene]}).
The two polymers indicated above are structurally similar and, in solution, are characterized by the same absorption spectrum with a peak in the proximity of 420 nm. The most significant difference is linked to the fact that the PPyPV is a base and, for this reason, is protonated by the HCl present in the solution in which the polymer is dissolved. The interaction with the carbon nanotubes favors the protonation process. The device obtained consists simply of two metal electrodes, between which are arranged the polymer/CNT blends (polymer-wrapped SWNTs), deposited by spin coating.
The above polymers and the carbon nanotubes are in electrical contact: in this way, it is possible to use radiation of appropriate wavelength for the purpose of modulating the electrical conductivity of the ropes of carbon nanotubes. The researchers carried out a series of measurements at different wavelengths for devices containing just ropes of CNTs, PmPV-wrapped/CNTs and PPyPV-wrapped/CNTs, and the current responses obtained are illustrated in FIGS. 6a, 6b, and 6c, respectively.
In particular, as emerges clearly from FIG. 6a, in the devices containing only CNTs no type of optically modulated response is detected. FIGS. 6b and 6c show, instead, the responses of the PmPV-wrapped/CNTs device with negative and positive electrical biasing, respectively. The variation of the output response is the same whatever the electrical biasing and is approximately 15-20% of the total current. In particular, there appears a photo-amplification of the current for positive biasing, i.e., the intensity of the total current in the device (dark current plus photogenerated current) increases, in absolute value, when the light at input is ON; on the other hand, for negative biasing there is a photorectification; consequently, in conditions of illumination the total current diminishes.
FIG. 7 illustrates, instead, the current response of a device with PPyPV-wrapped/CNTs. As may be noted, no effect is observed for negative biasing, while for positive biasing the effect of photo-amplification is substantially greater as compared to the previous case illustrated in FIGS. 6a, 6b, and 6c. 
Using structures of this type, it is possible to make optically controlled electrical switches. In fact, as is evident from FIGS. 6b and 6c, to a switching of the optical signal at input (light ON-light OFF) there corresponds a switching of the electrical signal at output.
The fundamental limits of the above devices are linked basically to the complexity of their implementation in integrated technology on account of the technique used for depositing the polymer/CNT blend. The spin-coating technique, in fact, renders it extremely difficult to define on the chip delimited areas on which to deposit the solution. The delineation of these areas would require a further process of controlled removal of the “spin-coated” solution, to be appropriately defined on the basis of the properties of the starting solution. Furthermore, spin coating does not enable a good control of the uniformity of thickness of the deposited film to be obtained.
Another important embodiment is linked to the low temperatures at which the devices have been made and tested (Top=4 K). From the literature, in fact, their behavior at room temperature is not known.
The purpose of embodiments is to provide an optically controlled electrical-switch device that will enable the drawbacks of the known devices described above to be overcome at least in part.